Substituent effects on the photochromic properties of benzothiophene-based derivatives

Article abstract of DOI:10.1002/chem.201500647

Five diarylethene photochromic derivatives, the structures of which incorporate a central benzothiophene unit, a left-hand thiazole group, and a right-hand benzothiophene group, have been prepared. The compound with a thiazole unit with no substituent on the reaction-center carbon atom reveals an unprecedented transformation upon light irradiation. When the 4-position of thiazole is protected by a methyl group, the compounds show high photosensitivity and photochromic properties. In this case, light irradiation affords new compounds with [5]helicene structures featuring the highest redshifted absorption maxima reported to date.

Full text of DOI:10.1002/chem.201500647

DOI: 10.1002/chem.201500647
Full Paper
&
Helical Structures
Substituent Effects on the Photochromic Properties of
Benzothiophene-Based Derivatives
Olivier Galangau,*[a, b] Takuyama Nakashima,[a] FranÅois Maurel,[c] and Tsuyoshi Kawai*[a, b]
Abstract: Five diarylethene photochromic derivatives, the
structures of which incorporate a central benzothiophene
unit, a left-hand thiazole group, and a right-hand benzothio-
phene group, have been prepared. The compound with
a thiazole unit with no substituent on the reaction-center
carbon atom reveals an unprecedented transformation upon
light irradiation. When the 4-position of thiazole is protected
by a methyl group, the compounds show high photosensi-
tivity and photochromic properties. In this case, light irradia-
tion affords new compounds with [5]helicene structures fea-
turing the highest redshifted absorption maxima reported to
date.
Introduction
cently, Guerchais et al. reported on CH direct arylation method-
ologies, involving benzothiophene as a central ring, which
demonstrates how famous and attractive this heterocycle has
become over the last few years.[14] Our group has also success-
fully employed this heterocyclic moiety as a centerpiece of a tri-
angular (so-called terarylene) photoquantitatively ring-cyclized
scaffold. Thanks to a series of intramolecular interactions that
stabilize the photoreactive conformation, high photocyclization
reactivity has been achieved.[13d] Herein, we wish to report on
some related compounds based on the benzothiophene-cen-
tered terarylene backbone. We developed five derivatives with
a pattern inspired by our previous results, comprising a left-
hand thiazole group and a right-hand benzothiophene core.
Minor changes in the substituents on the reaction centers
were performed to investigate their influence on the overall
photochromic properties. As shown herein, these slight pertur-
bations in the molecular structure and on its symmetry have
dramatic consequences for their photochromic features.
Diarylethenes (DAEs) and their aromatic analogous derivatives
have been the subject of intense research over the past two
decades. These classes of compounds undergo reversible pho-
toisomerization reactions between two stable forms: the ring-
open form with the hexatriene backbone and the ring-closed
form with the cyclohexadiene backbone.[1,2] Promising com-
pounds centered on perfluorocyclopentene have shown high
fatigue resistance and low thermal reactivity, which has made
them highly attractive for optoelectronics applications.[3] How-
ever, their syntheses usually require the use of expensive and
toxic octaperfluorocyclopentene, which inherently limits its
use.[4] Consequently, a wide variety of heteroaromatic central
bridging units have been introduced. Feringa et al. proposed
to replace the perfluoro group with a perhydro group,[5a]
whereas Hecht et al. suggested cyclopentene-related substan-
ces.[5b,6] Other groups have opted for aromatic rings, such as
thiophenes,[7] N-methylindole,[8] indenones,[9] imidazoles,[10]
thiazoles,[11] benzothiadiazoles,[12] or benzothiophene.[13] Re-
Results and Discussion
[a] Dr. O. Galangau, Dr. T. Nakashima, Prof. T. Kawai
International Collaborative Laboratory
Syntheses
Center for Frontier Science and Technology NAIST-CEMES
29, rue Jeanne Marvig, BP 94347, 31055 Toulouse (France)
E-mail: o-galangau@ms.naist.jp
Details of the syntheses of precursor compounds are given in
the Experimental Section. Compounds 3, 4, 5, and 6 were pre-
pared according to the synthetic route depicted in Scheme 1.
First, compounds 1a and 1b were converted into 2a and
2b, respectively, in very good yields, with Suzuki–Miyaura or-
ganometallic coupling conditions, as inspired by the work of
Guglielmetti et al.[15b] Indeed, we found that the use of an acti-
vated palladium catalyst was necessary because the coupling
involved two low-reactive positions (b–b coupling). From this
perspective, DME has been known to enhance the reactivity of
the palladium catalyst, probably due to strong chelation of the
metallic center by the two oxygen atoms. The corresponding
borylated reagents were prepared by conventional deprotona-
tion with nBuLi and exchange with tributyl borate. These rela-
tkawai@ms.naist.jp
[b] Dr. O. Galangau, Prof. T. Kawai
Graduate School of Materials Science
Nara Institute of Science and Technology, NAIST
8916-5 Takayama, Ikoma, Nara 630-0192 (Japan)
Fax: (+81)743-72-6179
[c] Prof. F. Maurel
Laboratoire Interfaces, Traitements, Organisation et
Dynamique des Systꢀmes (ITODYS)
CNRS UMR 7086, Universitꢁ Paris 7 - Paris Diderot
Sorbonne Paris Citꢁ, Bꢂtiment Lavoisier
15 rue Jean Antoine de B¨ıf, 75205 Paris Cedex 13 (France)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201500647.
Chem. Eur. J. 2015, 21, 1 – 13
1
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Scheme 1. Synthetic route to 2a, 2b, 3, 4, 5, and 6. DME=dimethoxyethane, PivOH=pivalic acid.
tively instable boronic esters were not isolated, but directly
combined with precursor C to afford derivatives 3 and 4 in ex-
cellent yields. Derivatives 5 and 6 were obtained from the CH
direct arylation coupling of intermediates 2a and 2b in good
yields. The use of the standard Fagnou conditions (tricyclohex-
ylphosphine combined with potassium carbonate in dimethy-
lacetamide) did not give satisfactory results, which justifies the
use of stronger conditions.[16] Finally, compound 9 was ob-
tained from a different synthetic pathway (Scheme 2).
thanks to modular, efficient, and optimized synthetic ap-
proaches.
Spectroscopic properties of derivatives 3 and 4
The spectroscopic properties of 3 and 4 were investigated in
THF at room temperature and the results are gathered in
Table 1. Both compounds showed similar characteristics, such
as a maximum of absorption located in the UV wavelength
range, at l=314 and 311 nm for 3 and 4, respectively.
Upon UV-light irradiation, these compounds underwent
a profound modification of their spectroscopic properties, as
shown in Figure 1 by the emer-
The CH direct arylation organometallic coupling with the
strongest Fagnou conditions afforded intermediate 7 in excel-
lent yield. A ratio of 3:1 thiazole/benzothiophene was necessa-
gence of new bands, especially
in the visible part, with maxima
located at l=525 nm for 3 and
l=543 nm for 4. These new ab-
sorption features can be safely
assigned to the appearance of
the corresponding closed-form
isomers. However, these new
bands thermally bleached a few
seconds after irradiation stopped
(dashed line in Figure 1), which
indicated subsequent degrada-
tion of the closed-form adducts
in both cases. These thermal
Scheme 2. Synthetic route to 7, 8, and 9. NBS=N-bromosuccinimide.
bleaching reactions were studied
in more detail by monitoring
1
ry to maximize the reaction yield. Bromination of the b position
on the benzothiophene core proceeded smoothly by using the
combination of NBS in THF from low to room temperature,
and 8 was employed without further purification. Eventually,
modified Suzuki–Miyaura conditions afforded the target mole-
cule 9 in excellent yield. Five sophisticated terarylene deriva-
tives were prepared in high yields in a minimum of steps
them with H NMR spectroscopy and HRMS. After UV irradia-
tion of the NMR spectroscopy tubes, the color faded quickly.
1
The initial H NMR spectrum of 3 showed quantitative conver-
sion of the open isomer, as proven by the complete disappear-
ance of the singlet at d=6.65 ppm, which was assigned to the
a-hydrogen atom on the thiazole ring, and the singlet at d=
2.33 ppm, which was assigned to the methyl group (see Fig-
&
&
Chem. Eur. J. 2015, 21, 1 – 13
www.chemeurj.org
2
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
nals (doublets, triplets). Krayushkin et al. recently reported dark
bleaching processes that occurred in DAE-related substances
possessing one hydrogen atom and a methyl group.[17] The 1D
NOE and 2D experiments (COSY, HMQC, and HMBC) were car-
ried out to determine the structure of the thermally generated
byproduct (see Figures S5 and S6 in the Supporting Informa-
tion). The HMQC experiment proved to be useful, since it
showed that the singlet signal at d=3.81 ppm of one proton
did not correlate with a carbon atom. The 1D NOE data also
demonstrated a strong correlation between the methyl signal
and the previous one (Figure 2), which meant that the two
groups were close in space. Based on these considerations, we
finally determined that, similarly to the report by Krayushkin
et al.,[17] the closed-form adduct underwent a [1,5]-proton
shift, followed by benzothiophene opening concomitant with
Table 1. Spectroscopic properties of 3 and 4 in THF.
[a]
lmax
emax
[Lmolꢀ1 cmꢀ1
lmax
emax
[Lmolꢀ1 cmꢀ1
]
Fo/c
Fc/o
[nm]
]
[nm]
3o
3c
4
314
–
311
–
31500
–
28500
–
–
525
–
–
–
–
–
0.2ꢁ0.1
–
–
–
–
–
0.3ꢁ0.1
4c
543
–
[a] Determined at l=313 nm.
Figure 2. 1H NMR spectra of 3 (lower spectrum=before irradiation, upper
spectrum=after irradiation). Inset: Correlations observed by the 1D NOE ex-
periment when the singlets at d=3.81 (solid line) and 2.42 ppm (dotted
line) were irradiated.
the appearance of a fully aromatized derivative, which provid-
ed the driving force. Similar conclusions were drawn for com-
pound 4 (see the Supporting Information). Assuming 1:1 pho-
toreactions for 3 and 4, we roughly evaluated their respective
ring cyclization quantum yields to be 0.2 and 0.3 for 3 and 4,
respectively, in THF at room temperature. These values are also
consistent with the population of the photoreactive conformer
determined by variable-temperature (VT) NMR spectroscopy
(see below) at the same temperature.
Figure 1. Absorption spectra of 3 (top) and 4 (bottom). Solid black line=op-
en form, solid gray line=photostationary state, and dashed line=after ther-
mal bleaching.
Briefly, thanks to VT-NMR spectroscopy and quantum yield
measurements, we clearly established that the photosensitivi-
ties of compounds 3 and 4 were rather low in THF at room
temperature. They also displayed intriguing and unusual pho-
tochemical behavior because their corresponding closed-form
isomers underwent rapid and quantitative bleaching, resulting
in full aromatization of the core and benzothiophene ring
cleavage.
ure S3 in the Supporting Information). Instead, the two initial
singlet signals were subjected to sizeable shifts: from d=2.33
to 2.42 ppm for the methyl signal and from d=6.65 to
3.81 ppm for the a-hydrogen atom. HRMS confirmed that the
reaction proceeded without loss of atoms for both derivatives
3 and 4 (not shown).
The dark-state transformation also induces splitting in the
aromatic area; the final spectrum displays nicely resolved sig-
Chem. Eur. J. 2015, 21, 1 – 13
www.chemeurj.org
3
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Spectroscopic properties of de-
rivatives 5, 6, and 9
Table 2. Spectroscopic properties of 5, 6, and 9 in THF.
[a]
[b]
lmax [nm]
e (lmax) [Lmolꢀ1 cmꢀ1
]
lmax [nm]
e (lmax) [Lmolꢀ1 cmꢀ1
]
Fo/c
Fc/o
c
The properties of compounds 5,
6 and 9 in THF were also investi-
gated and the results are gath-
ered in Table 2. These three com-
pounds display very similar spec-
troscopic properties, as expect-
5
6
9[c]
304
306
305
22200
25300
25400
540
562
544
13000
13800
–
0.73ꢁ0.04
0.74ꢁ0.04
–
0.09
<0.01
–
0.41
0.71
–
[a] Absolute values determined at l=313 nm. [b] Absolute values determined at l=540 nm. [c] We were
unable to evaluate the photoconversion, nor the quantum yields.
ed.
Upon
UV
irradiation
(Figure 3), their initial absorption spectra are deeply modified
and new bands arise in the visible part at l=540 (5), 562 (6),
and 544 nm (9). Notably, the OMe group of 6 induces a sizeable
redshift (22 nm) of the closed-form absorption maximum com-
pared with those of 5 or 9, as well as band broadening. These
photoprocesses seem to proceed in a clean way, as evidenced
by the appearance of isobestic points at 322 (5), 310 (6), and
324 nm (9).
1H NMR spectroscopic monitoring of their photoreactions
also supports this observation and has allowed us to evaluate
their respective photoconversion (see the Supporting Informa-
tion). However, evaluating the photochromic parameters of
compound 9 failed. None of the techniques (NMR spectroscop-
ic monitoring or irradiation followed by HPLC separation)
helped us to determine its photoconversion, probably because
of secondary processes that hamper its evaluation. Therefore,
only qualitatively spectroscopic results can be discussed for de-
rivative 9.
Their photoreversibility was also assessed by regenerating
the open isomer under visible-light irradiation. The three deriv-
atives showed poor fatigue resistances, even after one cycle,
since the absorption spectra observed after visible irradiation
did not match the initial ones (see the Supporting Information
for more details). Such a result is in accordance with previous
observations.[13b]
We also studied the thermal stability of the ring-closed
forms, 5c, 6c, and 9c. At high temperatures in toluene, all
three derivatives underwent general bleaching accompanied
by a monoexponential decrease in the absorbance. This sug-
gests that temperature-dependent degradation occurs, and
therefore, limits the stabilities of the closed forms. The Arrhe-
nius plots of each of sample yielded the corresponding activa-
tion energies, along with the calculated half-lives at 298 K (see
Table 3).
Compounds 5c and 6c feature similar activation energies of
100 and 105 kJmolꢀ1, respectively, whereas 9c shows the
smallest activation energy (76 kJmolꢀ1) along with the shortest
half-life (about 24 days). This suggests that substituting the
two reactive carbon sites enhances the stability of the thermal
closed-form adduct. Notably, upon going from 3c and 4c to
9c, that is, when switching the position of the hydrogen atom
from one ring to another, significant improvement of the sta-
bility of the closed isomer was achieved from few minutes to
several days.
Figure 3. UV/Visible spectra of 5 (top), 6 (middle), and 9 (bottom). Solid
black line=open isomers. Solid gray line=Photo-stationary state.
Ultimately, we evaluated the photochromic efficiencies of
these three derivatives by measuring their ring cyclization and
ring cycloreversion quantum yields in THF at room tempera-
ture. For both 5 and 6, we measured large improvements of
Fo/c compared with those of 3 and 4, respectively.
&
&
Chem. Eur. J. 2015, 21, 1 – 13
www.chemeurj.org
4
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Table 3. Calculated activation energies for the thermal degradation of
derivatives 5c, 6c, and 9c.
Table 5. Computed conformational exchange of 3a and 3b at 298 K.
Conformer
dNS
[ꢁ]
T1
[8]
T2
[8]
Population
AP [%] P [%]
DE
[kcalmolꢀ1
]
Ea [kJmolꢀ1
]
t1/2 [days] at 298 K
5c
6c
9c
100
105
76
103
144
24
AP
P
AP
P
2.93
2.94
3.02
2.93
ꢀ16
11
ꢀ25
6
111
69
111
70
44
–
36
–
–
56
–
3
4
0.14
0.33
64
Indeed, compounds 5 and 6 feature high values of around
70%, which makes them very photosensitive. These results
demonstrate 1) that substitution on the benzothiophene unit
has a small influence on the photosensitivity (no change be-
tween 5 and 6), and 2) the substitution of the reactive site on
the thiazole group plays a major role in stabilizing the photo-
reactive conformer in solution (e.g., 3 vs. 5). We suggest herein
play high photosensitivity, with population of the photoreac-
tive conformer of 70%. This fact strongly suggests that, since
the molecule is nonsymmetric, benzothiophene and thiazole
must play different roles in the conformation equilibrium (3 vs.
9). We propose herein that a CHꢀp interaction is established
between benzothiophene (p group) and the methyl group at-
tached to thiazole. This attractive interaction is then likely to
the existence of weak intramolecular interactions, such as CHꢀ stabilize the AP conformation, and therefore, drives the equi-
p bonding interactions, occurring between the methyl group
attached to thiazole and benzothiophene. Such an interaction
would partly drive the conformational equilibrium toward the
antiparallel (AP) conformer. Quantum yields and VT-NMR spec-
troscopy (see below) measurements suggest that the ring cyc-
lizations reflect, at least partly, the solution equilibrium state.
For Fc/o, the trend is reversed. Introducing a methoxy group
onto the molecular backbone quenches cycloreversion; this is
known for other DAEs.[18]
librium. Notably, these data match the previously mentioned
quantum yields well, and therefore, clearly demonstrate that,
for the whole compound family, the photocylizations are limit-
ed by their conformational exchange only, and the reactive
conformers undergo photon-quantitative cyclization.
To summarize, as opposed to previous results, thanks to VT-
NMR spectroscopy and photochemical measurements in THF,
we have demonstrated that adding a methyl group to the thia-
zole unit dramatically improves both the photosensitivity and
chemical stability in the dark for the closed-form adduct. How-
ever, when the two reactive carbon atoms are substituted by
groups other than hydrogen atoms, the stability is improved.
Moreover, it is worth mentioning that small variations of DG0
achieved by removing one methyl group induce remarkable
large discrepancies in the population ratio of reactive/nonreac-
tive isomers, which underlines the difficulty of rationally de-
signing highly photosensitive DAE systems.
VT-NMR spectroscopy data
NMR spectroscopy measurements have already proven to be
a valuable tool to examine the equilibrium between parallel (P)
and antiparallel (AP) conformers in solution that limits the ring
cyclization efficiency.[19] To determine whether the photocycli-
zation reactivity was limited by such an equilibrium, we moni-
tored specific chemical shifts in [D8]THF (see the Supporting In-
formation) as a function of the temperature (from 193 to
333 K). No splitting of signals was observed, which meant an
equilibrium was rapidly set up at each temperature. Results are
summarized in Table 4. Assuming fast equilibrium between the
two conformers (AP and P), we could calculate their respective
Boltzmann populations at 298 K (Table 5).
Theoretical investigations
Computations were performed on the five derivatives to gain
a better insight into their conformational, electronic, and UV/
Visible properties. We performed calculations by using the
Gaussian 09 (see the Supporting Information for a full refer-
ence) package by using a combination of the wB97XD func-
tional and the 6-31G(d) basis set for geometrical optimizations
(wB97XD/6-31G(d) calculations). Conformational analyses were
performed on the open isomer of compounds 3 and 5. TD-DFT
calculations were carried out at the wB97XD/6-311g(d,p) level
of theory. NMR spectroscopy predictions were performed at
the PBE0/6-311+ +g (2d,p) level. In the case of the non-pho-
toreactive conformers, geometries corresponding to an energy
minimum were reoptimized at the same level of theory prior
to NMR spectroscopy calculations. Conformational equilibria of
compounds 4, 6, and 9 were studied by considering the rele-
vant geometries established with scans performed on 3 and 5.
IR spectra of the optimized structures showed no imaginary
frequencies, which suggested that they corresponded to
energy minima. TD-DFT calculations of open-form isomers
Thus we obtained clear evidence that, if the methyl group is
positioned on the thiazole or benzothiophene moieties, the
population ratio is reversed. Although 3 and 4 show weak
photosensitivity (less than 30%), compounds 5, 6, and 9 dis-
Table 4. Calculated populations of the conformers and free Gibbs energy
at 298 K.
Population of conformers
DG0
[kcalmolꢀ1
]
AP [%]
P [%]
3
4
5
6
9
24
28
68
71
69
76
72
32
29
31
ꢀ0.7
ꢀ0.6
0.5
0.6
0.5
Chem. Eur. J. 2015, 21, 1 – 13
www.chemeurj.org
5
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
were achieved based on the geometries of photoreactive con-
formers.
Geometrical considerations on open isomers 5, 6, and 9
We proceeded in a similar way to compute the geometries of
5o, 6o, and 9o (Figure 5). For the three derivatives, the lowest
energetic configuration corresponds to an AP conformation
within both thiazole and benzothiophene groups to form large
torsion angles (T1 =ꢀ478, T2 =1218).
Geometrical considerations on 3 and 4
Geometrical optimizations were performed for each of the
compounds in vacuo (see the Supporting Information). In the
case of 3 and 4, the lowest energetic geometry corresponds to
a non-photoreactive one (see Figure 4).
Figure 4. Predicted geometries of 3 for the lowest (top) and highest
(bottom) energy configurations.
Figure 5. Predicted geometries of 9o for the lowest (top) and highest
(bottom) energy configurations.
Computations predicted almost coplanarity between the
thiazole and central benzothiophene groups, with a nitrogen–
sulfur distance (dNS) of 2.94 ꢁ for both compounds; this is in
the range of attractive NꢀS bonding interactions.[13d,20] In the
case of 3, the two torsion angles, T1 and T2, were calculated to
be 11 and 698, respectively. The former value indicates again
the coplanarity of both thiazole and benzothiophene. The
latter suggests that the end of benzothiophene must be dis-
connected from the rest of the molecule, probably because of
strong steric repulsions. From this perspective, it is clear that
the substituent at the end of benzothiophene has a minor role
to play into the relative stability of the geometry, since it is
projected outwards. In addition, computations allow us to re-
produce the conformational equilibrium by scanning the T2
value and fully optimizing the resulting geometry at each step.
Starting from the lowest energy geometry and rotating the
benzothiophene group counterclockwise, calculations predict
a second energy minimum, lying at slightly higher energy
(DE=0.14 kcalmolꢀ1), corresponding to an AP conformation.
This conformer shows torsion angles of T1 =ꢀ168 and T2 =
1118, along with a P/AP ratio of 56%/44%. Although calcula-
tions significantly overestimate these population ratios, one
should consider that computations reproduce the conforma-
tional equilibrium well. Similar conclusions can be drawn for
compound 4, for which the calculations predict a slightly im-
proved AP population (36%) with an energy gap of 0.33 kcal
Similar to 3, computation of the equilibrium was also per-
formed on compound 5o (see Figure S16 in the Supporting In-
formation). Three P conformations, P1, P2, and P3, were found.
The last two lie at high energies (2.24 kcalmolꢀ1 for P2 and
2.98 kcalmolꢀ1 for P3, as referenced to the lowest energy ge-
ometry), whereas P1 is only about 0.65 kcalmolꢀ1 higher than
the energy minimum. Therefore, conformations P2 and P3
should be barely populated at room temperature. Calculated
Boltzmann populations based on the computed energy gap
separating the AP and P conformers affords a 72%/28% ratio
for 9o, which is consistent with experimental VT-NMR spectros-
copy observations (Table 6). Similar conclusions can be drawn
for compounds 5o and 6o. As observed by a comparison be-
tween 5o, 6o, and 9o, the substituent located on the benzo-
thiophene has a limited impact on the equilibrium, as expect-
ed and observed in solution. However, going from 3 to 9 clear-
ly demonstrates the key role of the substituent located on
thiazole. Thus, we calculated the distance between the methyl
positioned on the thiazole and benzothiophene planes, in
their respective AP conformations (Table 6). The three deriva-
tives are predicted to show rather short distances of about
2.8 ꢁ, corresponding to typical distances involving CHꢀp bond-
ing interactions. However, since we were unable to provide ex-
perimental data for comparison, we suggest the existence of
interactions (attractive or repulsive) that force the molecular
scaffold to adopt the AP conformation.
molꢀ1
.
As a result, DFT calculations are in reasonable agreement
with the experimental observations. The existence of confor-
&
&
Chem. Eur. J. 2015, 21, 1 – 13
www.chemeurj.org
6
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
ever, the general trend observed
from VT-NMR spectroscopy data
is well reproduced by DFT calcu-
lations.
Table 6. Computed conformational exchange of 5, 6 and 9 at 298 K.
Conformer
dNS
T1
T2
[8]
Population
AP [%] P [%]
DE
[kcalmolꢀ1
dCHꢀp
[ꢁ]
[ꢁ]
[8]
]
These overall outcomes tend
to demonstrate that DFT calcula-
tions fairly reproduce the experi-
mental observations with quite
large discrepancies, depending
on the molecules under study.
AP
P
AP
P
AP
P
3.15
3.23
3.22
3.21
3.13
3.23
ꢀ49
53
ꢀ53
50
ꢀ47
52
12
62
122
58
121
54
75
–
79
–
72
–
–
25
–
21
–
28
2.75
–
2.81
–
2.75
–
5o
6o
9o
0.65
0.77
0.56
mational equilibria is predicted and the influence of substitu-
tion on the most stable molecular geometry is also demon-
strated; this is consistent with the VT-NMR data.
Table 7. Agreement of the NMR spectroscopy computations with the VT-
NMR data.
Recently, Yang et al. developed a new methodology to
reveal the existence of weak noncovalent interactions (NCIs).[21]
When comparing the three photoreactive conformers, a persis-
tent attractive NCI (green surfaces in Figure 6) occurred be-
Conformer
dVT [ppm]
dDFT [ppm]
R [%][a]
AP
P
AP
P
AP
P
1.42
2.58
1.76
2.64
1.42
2.56
1.67
1.99
1.79
2.05
1.57
1.73
17
ꢀ23
2
ꢀ22
10
5o
6o
9o
ꢀ32
[a] R=[(dDFTꢀdVT)]/dVT ꢂ100, in which dDFT and dVT are the chemical shifts
determined by DFT calculations and VT-NMR spectroscopy, respectively.
While the computed properties of compounds 3 and 4 poorly
match the experimental results, those of derivatives 5o, 6o,
and 9o are, on the contrary, well reproduced. For 5o, 6o, and
9o, computations show a clear combination of weak noncova-
lent nitrogen–sulfur interactions, steric repulsion, and weak
methyl/benzothiophene interactions that eventually result in
stabilization of the AP conformation. DFT calculations found
reasonably good agreement for the conformational equilibri-
um and predicted NMR spectra.
Figure 6. Visualization of the weak attractive NCIs in compound 9o.
tween the adjacent benzothiophene group and methyl at-
tached to the thiazole group, which was in agreement with
our previous DFT findings. Surprisingly, a second attractive in-
teraction between the two benzothiophene moieties (Figure 6
and Figure S17 in the Supporting Information) contributes to
stabilization of the conformation. Moreover, additional interac-
tions are foreseen between the substituent on benzothio-
phene and the thiazole group for compounds 5o and 6o (see
Figure S17 in the Supporting Information). These theoretical
observations tend to validate the concept of weak interactions,
such as CHꢀp attractive interactions, that also stabilize the
photoreactive conformer.
Geometrical considerations on closed isomers 5, 6, and 9
(5c, 6c, and 9c)
We also investigated the corresponding closed forms of 5, 6,
and 9 from the structural point of view (Figure 7).
In the closed form, the reactive carbon sites change their hy-
bridization from sp2 to sp3, as predicted by their adopted tetra-
hedral shape. Interestingly, the nitrogen and sulfur centers are
found at a distance of 3.15 ꢁ for all derivatives, which means
that the two still interact. Additionally, optimized structures
Calculated NMR spectroscopy considerations on 5, 6, and 9
open isomers
To further prove the correlation between calculations and ex-
perimental data, we computed the NMR spectra of both con-
formers for each of the samples 5o, 6o, and 9o (see the Sup-
porting Information for the methodology). The theoretical
chemical shift assigned to the methyl substituent located on
the thiazole ring was then calculated in vacuo at the PBE0/6-
311g(2d,p) level of theory.[19c]
As observed from the results in Table 7, calculations predict
slightly overestimated values for AP conformers, whereas the
chemical shifts of the P conformers are underestimated. How-
Figure 7. Predicted geometries for 5c, 6c, and 9c.
Chem. Eur. J. 2015, 21, 1 – 13
www.chemeurj.org
7
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
show clear characteristics of the helical backbone, with five
fused ring members connected to each other.[22] As a result,
a twist angle, T3, of 418 is formed between the two terminal
moieties. Although [5]helicenes or related substances are rare,
mainly due to their difficult preparation methods,[23] similar
values of helical curvature have been reported.[22b]
Table 9. Agreement of the vertical transitions with the UV/Visible data.
lexptl [nm]
lTD [nm]
Dl [nm]
5c
6c
9c
540
562
544
560
571
537
18
24
21
TD-DFT investigations into open isomers of 3, 4, 5, 6, and 9
The visible band was assigned to a pure HOMO!LUMO
transition in charge of the photochromic reaction. Indeed, al-
though the HOMOs present a typical bonding interaction, the
LUMOs display a characteristic antibonding one. Our molecular
systems compete well with previously published derivatives.[22-
To gain a better insight into the photochemical properties of
the compounds, we performed TD-DFT calculations in vacuo.
We present hereafter calculated vertical transitions of the five
open isomers of 3, 4, 5, 6, and 9 (see also the Supporting In-
formation). All open forms of the compounds display rather
usual photochromic properties. TD-DFT calculations predict
maxima of absorption located in the UV part of the electro-
magnetic spectra with a constant error relative to the experi-
mental data (see Table 8).
b,e–f]
Absorption maxima are all redshifted by about 40 to
60 nm, with a large increase in the molecular extinction coeffi-
cient values, although their photoconversions are fairly similar
to those reported by the Yokoyama group.[22b]
Conclusion
Table 8. Agreement of the vertical transitions with the UV/Visible data.
We prepared five photoactive compounds based on the terary-
lene model, incorporating a central benzothiophene moiety,
a left-hand thiazole group, and a right-hand benzothiophene
one. Two categories of derivatives could be suggested, de-
pending on whether the thiazole was substituted or not. When
it was unsubstituted, these derivatives featured complete pho-
totransformations into new aromatic derivatives functionalized
by a thiol group.
lexptl [nm]
lTD [nm]
Dl [nm]
3
4
5o
6o
9o
311
314
304
305
306
283
281/279
277
280
278
28
33/35
27
25
28
On the other hand, effective photochromic behavior was ob-
served in compounds with a thiazole unit substituted by
a methyl group. In other words, ring cyclization was character-
ized by high quantum yield values, moderate to high photo-
conversions, and afforded stable helical closed isomers, which
were the first of their kind to absorb far in the visible range.
On the basis of VT-NMR spectroscopy experiments and ring
cyclization quantum yield measurements in solution, we clearly
established that the photosensitivity of the open isomers was
strictly limited by their ground-state conformational exchange.
Theoretical computations afforded an understanding of the
conformational equilibrium by mimicking the experimental
data. They clearly revealed additional weak intramolecular in-
teractions to stabilize the AP conformer for 5, 6, and 9. On the
contrary, in scaffolds 3 and 4, such interactions vanished and
the photosensitivities were therefore lower. We are now focus-
ing our efforts on the enantioselective photogeneration of sim-
ilar helical closed isomer and, in the meantime, trying to im-
prove their overall stability.
The HOMOs exhibit conventional antibonding interactions
between the two carbon sites, whereas those of HOMO-
1 show opposite bonding interactions. As usually observed for
DAE and terarylene derivatives, the first transition is character-
ized by a small oscillating strength value, which is correlated
herein with the fact that the density flows from the two ben-
zothiophenes to the thiazole unit, resulting in poor overlap be-
tween the two series of molecular orbitals (charge-transfer ex-
citation). Surprisingly, the photochromic virtual molecular orbi-
tal, that is, the one populated after electronic excitation and
triggering the bond-forming step, does not correspond to the
LUMO, but to the LUMO+1. Indeed, as seen in the Supporting
Information, for all derivatives, the LUMO+1 presents an ade-
quate bonding interaction, whereas the LUMO shows the ab-
sence of density on one of the reactive carbon sites. Therefore,
the photochromically active transitions, herein S0!S2, are
mainly dominated by a density flow of HOMO!LUMO+1.
TD-DFT investigations into closed isomers of 5, 6, and 9 (5c,
6c, and 9c)
Experimental Section
Vertical transitions of derivatives 5c, 6c, and 9c were comput-
ed at the same level of theory. All show identical photophysical
behavior (Table 9), with theoretical absorption spectra match-
ing those obtained experimentally within reasonable error
values.
Chemicals/purification/analyses
Chemicals were purchased from Tokyo Chemical Industries, Wako
Chemicals, and Sigma Aldrich and were used without any further
treatment. Dry solvents were purchased from Nacalai Tesque and
Sigma Aldrich, and kept under a nitrogen atmosphere. Spectro-
scopic-grade solvents were purchased from Wako chemicals. 1H
Note that the calculations reproduce well the redshift of the
absorption maximum in 6c relative to the rest of the family.
&
&
Chem. Eur. J. 2015, 21, 1 – 13
www.chemeurj.org
8
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
and 13C NMR spectra were recorded on
a JEOL JNM-AL-300
Synthesis of 2b
(300 MHz) and JEOL JNM-ECP400 (400 MHz) spectrometers. The 2D
NMR spectra were recorded on a JEOL ECA-600 MHz spectrometer.
Chemical shifts were referenced to the residual solvent signals.
Mass spectra were measured with a JML-700 (EI) mass spectrome-
ter. ATR-IR spectra were recorded on a JASCO FT/IR-4200 spectrom-
eter with an ATR PRO410-S attachment. Melting points were mea-
sured with a capillary MEL-TEMP apparatus. Purifications by flash
chromatography were performed on a YAMAZEN W-Prep 2XY
system. Compounds studied by spectroscopic measurements of
any kind were first purified by either HPLC (normal phase: HITACHI,
LaChrom ELITE, L-2400 as detector,L-2130 as pump, D-2500 as inte-
grator with a Nacalai Tesque, COSMOSIL 55 L-II, 20ꢂ250 mm
packed column and reversed phase: LaChrom ELITE, L-2455 as de-
tector, L-2130 as pump with a Nacalai Tesque COSMOSIL 5C18
201Dꢂ250 mm packed column) or GPC (8391 NEXT HPLC from
Japan Analysis Industry).
A Schlenk tube equipped with a magnetic stirring bar and capped
with a rubber septum was charged with compound 1b (1.37 g,
5.6 mmol, 1 equiv), commercially available benzo[b]thiophen-3-yl-
boronic acid (1.5 g, 8.4 mmol, 1.5 equiv), and anhydrous sodium
carbonate (0.893 g, 8.4 mmol, 1.5 equiv). DME (140 mL) was then
added and the suspension was evacuated/backfilled with nitrogen
several times. Tetrakis(triphenylphosphine) palladium(0) (0.195 g,
3 mol%) was added under a stream of nitrogen followed by dis-
tilled water (70 mL), and the monophasic mixture was evacuated/
backfilled again. The solution was heated at 75–858C over a period
of 24 h (TLC monitoring) and allowed to cool to RT. After diluting
the mixture with water, the aqueous layer (purple) was extracted
with ethyl acetate several times. The organic phases were com-
bined, dried over anhydrous magnesium sulfate, filtered, and the
solvent was removed. The crude product was purified by column
chromatography (silica, hexane to mixture hexane/CHCl3, 9/1, v/v)
1
to give 2b as a white solid (1.3 g, 78%). M.p. 147–1498C; H NMR
(300 MHz, CDCl3): d=3.93 (s, 3H), 7.22–7.39 (m, 5H), 7.49 (s, 1H),
7.57 (d, 1H, J=6 Hz), 7.71–7.74 (m, 2H), 7.92 ppm (d, 1H, J=6 Hz);
13C NMR (75 MHz, CDCl3): d=61.6, 109.74, 121.8, 122.2, 122.7,
122.9, 123.6, 123.9, 124.3, 124.7. 125.5, 125.6, 128.7, 130.9, 138.4,
Photophysical studies
Absorption spectra were recorded with spectrophotometers
(JASCO V-670 and V-660). A Compact Asahi spectra xenon light
source was used as an irradiation source. Light with an appropriate
wavelength was obtained by being passed through additional fil-
ters. Absolute ring cyclization quantum yield values were obtained
on a Shimadzu QYM-01 setup. Ring cyclization reaction quantum
yields of instable derivatives were measured relative to the refer-
ence compound 1,2-bis(2-methylbenzo[b]thiophene-3-yl)perhexa-
fluorocyclopentene, the value of which was 0.35 under l=313 nm
light irradiation in hexane (see the Supporting Information for a ref-
erence), by determining their respective photoconversion under
identical irradiation conditions. The relative quantum yields were
obtained from the relationship expressed in the Supporting Infor-
mation.
~
138.5, 139.9, 161.1 ppm; IR: n=3090, 3051, 2922, 2851, 2831, 1594,
1577, 1544, 1498, 1456, 1420, 1319, 1266, 1256, 1224, 1152, 1070,
1011, 766, 749, 726 cmꢀ1; HRMS (EI): m/z calcd for C17H12OS2 [M+]:
296.0330; found: 296.0330.
Synthesis of 3
A
flame-dried Schlenk tube was charged with 2a (500 mg,
1.8 mmol, 1.5 equiv) and dry THF (approximately 30 mL). The solu-
tion was cooled to ꢀ788C before a solution of nBuLi in hexane
(1.6m, 1.79 mL, 1.6 equiv) was added dropwise by means of a sy-
ringe. The solution turned brown and after 45 min at that tempera-
ture neat tributylborate (0.481 mL, 1.8 mmol, 1.5 equiv) was added.
The low-temperature bath was removed and the solution was al-
lowed to warm to RT over a period of 1 h or more. In the mean-
time, the solution turned red and then pale orange. Finally, the
content was transferred by means of a syringe to a second Schlenk
tube containing 4-bromothiazole (286 mg, 1.2 mmol, 1 equiv),
sodium carbonate (189 mg, 1.8 mmol, 1.5 equiv), and tetrakis(tri-
phenylphosphine) palladium(0) (41 mg, 3 mol%) dissolved into
a degassed 1:1 mixture of DME/water mixture. The combined solu-
tion was warmed at 65–758C until TLC showed reaction comple-
tion. The crude product was allowed to cool to RT, diluted with
water, and extracted with ethyl acetate. The organic phases were
combined, dried over anhydrous magnesium sulfate, filtered, and
the solvent was removed. Purification by flash chromatography
(silica, hexane/CHCl3, 9:1, v/v) afforded 3 as a white powder
(470 mg, 90%). M.p. 70–758C; 1H NMR (300 MHz, CDCl3): d=2.33
(s, 3H), 6.53 (s, 1H), 7.19–7.46 (m, 9H), 7.87–7.98 ppm (m, 4H);
13C NMR (75 MHz, CDCl3): d=14.5, 114.3, 119.2, 122.2, 122.3, 122.4,
122.7, 123.2, 124.2, 124.4, 124.5, 124.6, 124.9, 126.0, 126.6, 127.4,
128.9, 130.2, 133.1, 138.7, 139.0, 139.1, 139.3, 140.8, 149.4,
Synthesis of 2a
A Schlenk tube equipped with a magnetic stirring bar and capped
with a rubber septum was charged with 1a (1.28 g, 5.6 mmol,
1 equiv), commercially available benzo[b]thiophen-3-ylboronic acid
(1.5 g, 8.4 mmol, 1.5 equiv), and anhydrous sodium carbonate
(0.893 g, 8.4 mmol, 1.5 equiv). DME (169 mL) was then added and
the suspension was evacuated/backfilled with nitrogen several
times. Tetrakis(triphenylphosphine) palladium(0) (0.195 g, 3 mol%)
was added under a stream of nitrogen followed by distilled water
(84 mL), and the monophasic mixture was evacuated/backfilled
again. The solution was heated at 75–858C over a period of 24 h
(TLC monitoring) and was allowed to cool to RT. After diluting the
mixture with water, the aqueous layer (purple) was extracted with
ethyl acetate several times. The organic phases were combined,
dried over anhydrous magnesium sulfate, filtered, and the solvent
was removed. The crude product was purified by column chroma-
tography (silica, hexane) to give 2a as a white solid (1.41 g, 90%).
~
166.8 ppm; IR: n=3116, 3054, 2847, 1505, 1482, 1456, 1437, 1312,
1250, 1172, 1152, 1014, 965, 782, 759, 733 cmꢀ1; HRMS (EI): m/z
calcd for C26H17NS3 [M+]: 439.0523; found: 439.0511.
1
M.p. 65–708C; H NMR (300 MHz, CDCl3): d=2.44 (s, 3H), 7.22–7.44
(m, 6H), 7.54 (s, 1H), 7.81 (d, 1H, J=6 Hz), 7.95 ppm (d, 1H, J=
6 Hz); 13C NMR (75 MHz, CDCl3): d=14.8, 122.0, 122.6, 122.8, 123.2,
123.4, 123.9, 124.1, 124.2, 124.3, 124.5, 124.6, 124.7, 125.6, 127.5,
Synthesis of 4
~
130.7, 138.2, 138.3, 138.8, 140.0, 140.4 ppm; IR: n=3086, 3056,
1452, 1422, 1344, 1314, 1253, 1180, 1153, 1019, 797, 757, 727 cmꢀ1
HRMS (EI): m/z calcd for C17H12S2 [M+]: 280.0380; found: 280.0372.
A flame-dried Schlenk tube was charged with 2b (500 mg,
1.7 mmol, 1.5 equiv) and dry THF (approximately 20 mL). The solu-
tion was cooled to ꢀ788C before a solution of nBuLi in hexane
;
Chem. Eur. J. 2015, 21, 1 – 13
www.chemeurj.org
9
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
(1.6m, 1.70 mL, 1.6 equiv) was added dropwise by means of a sy-
ringe. The solution turned brown and after 45 min at that tempera-
ture neat tributylborate (0.455 mL, 1.7 mmol, 1.5 equiv) was added.
The low-temperature bath was removed and the solution was al-
lowed to warm to RT over a period of 1 h or more. In the mean-
time, the solution turned red and then pale orange. Finally, the
content was transferred by means of a syringe to a second Schlenk
tube containing 4-bromothiazole (270 mg, 1.1 mmol, 1 equiv),
sodium carbonate (179 mg, 1.7 mmol, 1.5 equiv), and tetrakis(tri-
phenylphosphine) palladium(0) (39 mg, 3 mol%) dissolved in a de-
gassed 1:1 mixture of DME/water. The combined solution was
warmed at 65–758C until TLC showed reaction completion. The
crude mixture was allowed to cool to RT, diluted with water, and
extracted with ethyl acetate. The organic phase was combined,
dried over anhydrous magnesium sulfate, filtered, and the solvent
was removed. Purification by flash chromatography (silica, hexane/
CHCl3, 9/1, v/v) afforded 4 as a white powder (460 mg, 90%). M.p.
TLC monitoring showed completion it was partitioned between an
aqueous layer and ethyl acetate. The biphasic system was filtered
through a pad of Celite and the organic phase was separated,
dried over anhydrous magnesium sulfate, and filtered. After remov-
al of the solvents, the crude product was purified by flash chroma-
tography (silica, hexane/CHCl3) to afford 6 as a white solid
(270 mg, 85%). M.p. 57–608C; 1H NMR (300 MHz, CDCl3): d=1.92
(s, 3H), 3.77 (s, 3H), 7.14–7.42 (m, 8H), 7.53 (d, 1H, J=6 Hz), 7.56
(d, 1H, J=6 Hz), 7.85–7.93 ppm (m, 3H); 13C NMR (75 MHz, CDCl3):
d=12.3, 61.4, 109.8, 122.0, 122.2, 122.7, 123.9, 124.1, 124.7, 124.8,
126.3, 127.1, 128.8, 128.9, 129.7, 130.9, 131.7, 133.6, 135.5, 137.6,
~
139.2, 139.8, 146.0, 161.1, 163.9 ppm; IR: n=3058, 2835, 1597,
1575, 1542, 1512, 1493, 1460, 1449, 1423, 1312, 1266, 1155, 1139,
1017, 756, 726 cmꢀ1; HRMS (EI): m/z calcd for C27H19NOS3 [M+]:
469.0629; found: 469.0619.
Synthesis of 7
1
145–1488C; H NMR (300 MHz, CDCl3): d=3.87 (s, 3H), 6.80 (s, 1H),
7.14–7.44 (m, 9H), 7.76–7.98 ppm (m, 4H); 13C NMR (75 MHz,
CDCl3): d=61.7, 109.2, 114.6, 122.0, 122.2, 122.3, 123.2, 123.3,
124.2, 124.3, 124.8, 125.0, 126.6, 128.9, 130.1, 131.3, 133.2, 136.6,
A Schlenk tube was charged with benzothiophene (791 mg,
6 mmol, 3 equiv), D (500 mg, 2 mmol, 1 equiv), PivOH (60 mg,
30 mol%), P(tBu)2MeHBF4 (49 mg, 10 mol%), anhydrous cesium car-
bonate (1.28 g, 4 mmol, 2 equiv), and palladium(II) acetate (22 mg,
5 mol%) under a stream of nitrogen. Dry mesitylene was added by
means of a syringe (3 mL) and the suspension was evacuated/back-
filled with nitrogen several times. The mixture was heated at
1408C (at high temperature the reactants solubilize and the reac-
tion starts). The mixture was allowed to cool to RT and after TLC
monitoring showed completion it was partitioned between an
aqueous layer and ethyl acetate. The biphasic system was filtered
through a pad of Celite and the organic phase was separated,
dried over anhydrous magnesium sulfate, and filtered. After remov-
al of solvent, the crude product was purified by column chroma-
tography (silica, hexane/CHCl3, 1:4, v/v) to afford 7 as a white solid
(550 mg, 91%). M.p. 95–978C; 1H NMR (300 MHz, CDCl3): d=2.70
(s, 3H), 7.31–7.58 (m, 5H), 7.77 (s, 1H), 7.79–7.97 ppm (m, 4H);
13C NMR (75 MHz, CDCl3): d=13.0, 121.4, 122.1, 123.6, 124.3, 126.3,
128.8, 129.0, 129.9, 133.3, 138.6, 139.7, 140.3, 145.8, 163.6 ppm; IR:
~
137.5, 138.9, 140.7, 149.5, 161.6, 166.6 ppm; IR: n=3133, 3057,
3010, 1598, 1573, 1551, 1532, 1502, 1486, 1464, 1447, 1434, 1420,
1357, 1316, 1282, 1262, 1233, 1183, 1155, 1016, 1002, 763, 743,
724 cmꢀ1; HRMS (EI): m/z calcd for C26H18NOS3 [M+H]+: 456.0551;
found: 456.0552.
Synthesis of 5
A Schlenk tube was charged with 2a (250 mg, 0.9 mmol, 1 equiv),
D
(227 mg, 0.9 mmol, 1 equiv), PivOH (27 mg, 30 mol%),
P(tBu)2MeHBF4 (22 mg, 10 mol%), anhydrous cesium carbonate
(582 mg, 1.8 mmol, 2 equiv), and palladium(II) acetate (10 mg,
5 mol%) under a stream of nitrogen. Dry mesitylene was added by
means of a syringe (3 mL) and the suspension was evacuated/back-
filled with nitrogen several times. The mixture was heated at
1408C (at high temperature the reactants solubilize and the reac-
tion starts). The mixture was allowed to cool to RT and after TLC
monitoring showed completion it was partitioned between an
aqueous layer and ethyl acetate. The biphasic system was filtered
through a pad of Celite and the organic phase was separated,
dried over anhydrous magnesium sulfate, and filtered. After remov-
al of the solvent, the crude product was purified by flash chroma-
tography (silica, hexane/CHCl3) to afford 5 as a white solid
~
n=3048, 3018, 2852, 1493, 1455, 1440, 1311, 1244, 1197, 1152,
1125, 1067, 1015, 976, 826, 762, 744, 725 cmꢀ1; HRMS (EI): m/z
calcd for C18H13NS2 [M+]: 307.0489; found: 307.0497.
Synthesis of 8
A brown-glass, round-bottomed flask was charged with 5 (400 mg,
1.3 mmol, 1 equiv) and THF (13 mL). NBS (255 mg, 1.4 mmol,
1.1 equiv) was added in one portion at 08C and the mixture was al-
lowed to slowly warm to RT overnight. Water was added and the
suspension was extracted with ethyl acetate. The organic phase
was washed with an approximately 1m aqueous solution of NaOH,
then with water, dried over anhydrous magnesium sulfate, and fi-
nally filtered. The solvent was removed and the crude product was
used without further purification (500 mg, 99%). M.p. 88–908C;
1H NMR (300 MHz, CDCl3): d=2.51 (s, 3H), 7.38–7.46 (m, 5H), 7.79–
7.95 (m, 4H); 13C NMR (75 MHz, CDCl3): d=13.2, 108.4, 122.2, 123.6,
125.1, 125.7, 126.3, 128.8, 129.9, 132.4, 133.3, 133.6, 138.1, 138.6,
1
(250 mg, 62%). M.p. 145–1488C; H NMR (300 MHz, CDCl3): d=1.82
(s, 3H), 2.22 (s, 3H), 7.18–7.43 (m, 11H), 7.77–7.96 ppm (m, 4H);
13C NMR (75 MHz, CDCl3): d=12.3, 14.9, 121.9, 122.2, 123.0, 123.7,
123.9, 124.1, 124.2, 124.8, 126.3, 127.3, 128.7, 128.8, 129.7, 131.9,
133.5, 136.0, 138.2, 138.8, 139.5, 139.7, 139.8, 145.6, 164.1 ppm; IR:
~
n=3054, 2946, 1948, 1512, 1496, 1453, 1430, 1308, 1289, 1263,
1250, 1171, 1155, 1142, 1021, 971, 759, 740, 726 cmꢀ1; HRMS (EI):
m/z calcd for C26H19NS3 [M+]: 453.0680; found: 453.0681.
Synthesis of 6
~
A Schlenk tube was charged with 2b (200 mg, 0.7 mmol, 1 equiv),
144.1, 164.7 ppm; IR: n=3058, 2920, 2847, 1505, 1489, 1488, 1460,
D
(172 mg, 0.7 mmol, 1 equiv), PivOH (21 mg, 30 mol%),
1437, 1306, 1277, 1250, 1162, 1070, 1017, 968, 814, 749, 719 ppm;
HRMS (EI): m/z calcd for C18H1NBrS2 [M+]: 384.9595; found
384.9606.
P(tBu)2MeHBF4 (17 mg, 10 mol%), anhydrous cesium carbonate
(441 mg, 14 mmol, 2 equiv), and palladium(II) acetate (8 mg,
5 mol%) under a stream of nitrogen. Dry mesitylene was added by
means of a syringe (2.3 mL) and the suspension was evacuated/
backfilled with nitrogen several times. The mixture was heated at
1408C (at high temperature the reactants solubilize and the reac-
tion starts). The mixture was allowed to cool down RT and after
Synthesis of 9
A Schlenk tube equipped with a magnetic stirring bar and capped
with a rubber septum was charged with 6 (361 mg, 0.9 mmol,
&
&
Chem. Eur. J. 2015, 21, 1 – 13
www.chemeurj.org
10
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
[5] a) L. N. Lucas, J. J. D. de Jong, J. H. van Esch, R. M. Kellogg, B. L. Feringa,
Eur. J. Org. Chem. 2003, 155–166; b) R. Gçstl, S. Hecht, Angew. Chem.
Int. Ed. 2014, 53, 8784–8787.
[6] R. Gçstl, B. Kobin, L. Grubert, M. Pꢅtzel, S. Hecht, Chem. Eur. J. 2012, 18,
14282–14285.
1 equiv), commercially available benzo[b]thiophen-3-yl boronic
acid (250 mg, 1.4 mmol, 1.5 equiv), and anhydrous potassium car-
bonate (194 mg, 1.4 mmol, 1.5 equiv). DME (23 mL) was then
added and the suspension was evacuated/backfilled with nitrogen
several times. Tetrakis(triphenylphosphine) palladium(0) (32 mg,
3 mol%) was added under a stream of nitrogen, followed by dis-
tilled water (12 mL), and the monophasic mixture was evacuated/
backfilled again. The solution was heated at 60–758C over a period
of 24 h (TLC monitoring) and allowed to cool to RT. After diluting
the mixture with water, the aqueous layer (purple) was extracted
with ethyl acetate several times. The organic phases were com-
bined, dried over anhydrous magnesium sulfate, filtered, and the
solvent was removed. The crude product was purified by flash
column chromatography (silica, hexane/CHCl3) to afford 9 as
[7] For recent examples, see: a) T. Nakashima, K. Atsumi, S. Kawai, T. Naka-
gawa, Y. Hasegawa, T. Kawai, Eur. J. Org. Chem. 2009, 3112–3218;
b) H. L. Wong, W.-T. Wong, V. W.-W. Yam, Org. Lett. 2012, 14, 1862–
1865; c) C. T. Poon, V. W.-W. Lam, J. Am. Chem. Soc. 2011, 133, 19622–
19625; d) J. C.-H. Chan, W.-H. Lam, H. L. Wong, W. T. Wong, V. W.-W. Yam,
Angew. Chem. Int. Ed. 2013, 52, 11504–11508; Angew. Chem. 2013, 125,
11718–11722.
[8] S. Fukumoto, T. Nakashima, T. Kawai, Dyes Pigm. 2012, 92, 868–871.
[9] a) K. Morinaka, T. Ubukata, Y. Yokoyama, Org. Lett. 2009, 11, 3890–3893;
b) M. Kose, E. Orhan, K. Suzuki, A. Tutar, C. Uenlue, Y. Yokoyama, J. Pho-
tochem. Photobiol. A 2013, 257, 50–53; c) H. Ogawa, K. Takagi, T, Uba-
kata, A. Okamoto, N. Yonezawa, S. Delbaere, Y. Tokoyama, Chem.
Commun. 2012, 48, 11838–11840; d) F. G. Erko, J. Berthet, H. Ogawa, Y.
Yokoyama, S. Delbaere, Tetrahedron Lett. 2013, 54, 6366–6369.
[10] a) B. M. Neilson, V. M. Lynch, C. W. Bielawski, Angew. Chem. Int. Ed. 2011,
50, 10322–10326; Angew. Chem. 2011, 123, 10506–10510; b) B. M. Neil-
son, C. W. Bielawski, J. Am. Chem. Soc. 2012, 134, 12693–12699; c) V. W.-
W. Yam, J. K.-W. Lee, C.-C. Ko, N. Zhu, J. Am. Chem. Soc. 2009, 131, 912–
913; d) P. H.-M. Lee, C.-C. Ko, N. Zhu, V. W.-W. Yam, J. Am. Chem. Soc.
2007, 129, 6058–6059; e) T. Nakashima, M. Goto, S. Kawai, T. Kawai, J.
Am. Chem. Soc. 2008, 130, 14570–14575.
1
a white solid (340 mg, 83%). M.p. 708C; H NMR (300 MHz, CDCl3):
d=2.44 (s, 3H), 7.22–7.44 (m, 6H), 7.54 (s, 1H), 7.81 (d, 1H, J=
6 Hz), 7.95 ppm (d, 1H, J=6 Hz); 13C NMR (75 MHz, CDCl3): d=14.8,
122.0, 122.6, 122.8, 123.2, 123.4, 123.9, 124.1, 124.2, 124.3, 124.5,
124.6, 124.7, 125.6, 127.5, 130.7, 138.2, 138.3, 138.8, 140.0,
~
140.4 ppm; IR: n=3058, 2913, 1575, 1456, 1437, 1420, 1312, 1286,
1250, 1237, 1155, 1070, 1021, 972, 756, 726 cmꢀ1; HRMS (EI): m/z
calcd for C26H17NS3 [M+]: 439.0523; found: 439.0520.
[11] a) T. Nakashima, K. Atsumi, S. Kawai, T. Nakagawa, Y. Hasegawa, T. Kawai,
Eur. J. Org. Chem. 2007, 3212–3218; b) K. Ouhenia-Ouadahi, R. Yasukuni,
P. Yu, G. Laurent, C. Pavageau, J. Grand, J. Guꢆrin, A. Lꢆaustic, N. Fꢆlidj,
J. Aubard, K. Nakatani, R. Mꢆtivier, Chem. Commun. 2014, 50, 7299–
7302; c) T. Nakagawa, K. Atsumi, T. Nakashima, Y. Hasegawa, T. Kawai,
Chem. Lett. 2007, 36, 372–373; d) T. Nakagawa, Y. Hasegawa, T. Kawai, J.
Phys. Chem. A 2008, 112, 5096–5103; e) Y. Kutsunugi, S. Kawai, T. Naka-
shima, T. Kawai, New. J. Chem. 2009, 33, 1368–1373; f) H. Nakagawa, S.
Kawai, T. Nakashima, T. Kawai, Org. Lett. 2009, 11, 1475–1478; g) M. Ta-
guchi, T. Nakagawa, T. Nakashima, T. Kawai, J. Mater. Chem. 2011, 21,
17425–17432; h) M. Taguchi, T. Nakagawa, T. Nakashima, C. Adachi, T.
Kawai, Chem. Commun. 2013, 49, 6373–6375.
Acknowledgements
We thank the Japan Society for Promotion of Science (JSPS)
and the Centre National de la Recherche Scientifique (CNRS)
for their financial support. A part of the present study has
been supported with a Grant-in-Aid for Scientific Research on
Priority-Area, Photosynergetics.
[12] a) W. Zhu, Y. Yang, R. Mꢆtivier, Q. Zhang, R. Guillot, Y. Xie, H. Tian, K. Na-
katani, Angew. Chem. Int. Ed. 2011, 50, 10986–10990; Angew. Chem.
2011, 123, 11178–11182; b) W. Li, C. Jiao, X. Li, Y. Xie, K. Nakatani, H.
Tian, W. Zhu, Angew. Chem. Int. Ed. 2014, 53, 4603–4607.
Keywords: arylation
photochromism · substituent effects
· helical structures · isomerization ·
[13] a) S. Chen, Y. Yang, Y. Wu, H. Tian, W. Zhu, J. Mater. Chem. 2012, 22,
5486–5494; b) Y.-C. Jeong, C. Gao, I. S. Lee, S. I. Yang, K.-H. Ahn, Tetrahe-
dron Lett. 2009, 50, 5288–5290; c) S. Cheng, L.-J. Chen, H.-B. Yang, H.
Tian, W. Zhu, J. Am. Chem. Soc. 2012, 134, 13596–13599; d) S. Fukumo-
to, T. Nakashima, T. Kawai, Angew. Chem. Int. Ed. 2011, 50, 1565–1568;
Angew. Chem. 2011, 123, 1603–1606.
[14] K. Yuan, J. Boixel, A. Chantzis, D. Jacquemin, V. Guerchais, H. Doucet,
Chem. Eur. J. 2014, 20, 10073–10083.
[15] a) R. Sanz, V. Guilarte, E. Hernando, A. M. Sanjuꢇn, J. Org. Chem. 2010,
75, 7443–7446; b) A. Heynderickx, A. Samat, R. Guglielmetti, Synthesis
2002, 213–216.
[1] a) Photochromism, Wiley-Interscience, New-York, 1971; b) H. Dꢃrr, H.
Bouas-Laurent in Photochromism, Molecules and Systems, Studies in Or-
ganic Chemistry 40, Elsevier, Amsterdam, 1990.
[2] a) V. Balzani, A. Credi, M. Venturi in Molecular Devices and Machines: A
Journey into the Nano World, Wiley-VCH, Weinheim, 2003; b) J. A. De-
laire, K. Nakatani, Chem. Rev. 2000, 100, 1817–1816; c) Y. Hasegawa, T.
Nakagawa, T. Kawai, Coord. Chem. Rev. 2010, 254, 2643–2651; d) M. Irie,
Chem. Rev. 2000, 100, 1684–1716; e) Y. Yokoyama, Chem. Rev. 2000,
100, 1717–1740.
[3] a) M. Irie, T. Fukaminato, T. Sasaki, N. Tamai, T. Kawai, Nature 2002, 420,
759; b) V. Guerchais, L. Ordonneau, H. Le Bozec, Coord. Chem. Rev. 2010,
254, 2533–2545; c) H. Tian, S. Yang, Chem. Soc. Rev. 2004, 33, 85–97;
d) G. Szalꢄki, G. Sevez, J. Berthet, J. L. Pozzo, S. Delbaere, J. Am. Chem.
Soc. 2014, 136, 13510–13513; e) S. Takami, T. Kawai, M. Irie, Eur. J. Org.
Chem. 2002, 3796–3800; f) H. Nishi, T. Namari, S. Kobatake, J. Mater.
Chem. 2011, 21, 17249–17258; g) H. Shoji, D. Kitagawa, S. Kobatake,
Chem. Commun. 2014, 38, 933–941; h) Y. C. Jeong, S. I. Yang, K. H. Ahn,
E. Kim, Chenm. Comm. 2005, 2503–2505; i) K. Uno, H. Niikura, M. Mori-
moto, Y. Ishibashi, H. Miyasaka, M. Irie, J. Am. Chem. Soc. 2011, 133,
13558–13564; j) H. Nitadori, L. Ordonneau, J. Boixel, D. Jacquemin, A.
Boucekkine, A. Singh, M. Akita, I. Ledoux, V. Guerchais, H. Le Bozec,
Chem. Commun. 2012, 48, 10395–10397; k) F. Meng, Y.-M. Hervault, Q.
Shao, B. Hu, L. Norel, S. Rigaut, X. Chen, Nat. Commun. 2013, 5, 3023–
3032; l) M. Irie, T. Fukaminato, K. Matsuda, S. Kobatake, Chem. Rev.
2014, 114, 12174–12277.
[16] B. Liꢆgault, I. Petrov, S. I. Gorelsky, K. Fagnou, J. Org. Chem. 2010, 75,
1047–1060.
[17] A. G. Lvov, V. Z. Shirinian, V. V. Kachala, A. M. Kavun, I. V. Zavarin, M. M.
Krayushkin, Org. Lett. 2014, 16, 4532–4535.
[18] a) S. Nakamura, T. Kobayashi, A. Takata, K. Uchida, Y. Asano, A. Muraka-
mi, A. Goldberg, D. Guillaumont, S. Yokojima, S. Kobatake, M. Irie, J.
Phys. Org. Chem. 2007, 20, 821–829; b) K. Morimitsu, K. Shibata, S. Ko-
batake, M. Irie, J. Org. Chem. 2002, 67, 4574–4578; c) T. Fukaminato, T.
Umemoto, Y. Iwata, S. Yokojima, M. Yoneyama, S. Nakamura, M. Irie, J.
Am. Chem. Soc. 2007, 129, 5932–5938; d) S. Kobatake, Y. Matsumoto, M.
Irie, Angew. Chem. Int. Ed. 2005, 44, 2148–2151; Angew. Chem. 2005,
117, 2186–2189; e) A. de Meijere, Z. Ligang, V. N. Belov, M. Bossi, N.
Noltmeyer, W. S. Hell, Chem. Eur. J. 2007, 13, 2503–2516; f) T. Fukamina-
to, T. Sasaki, T. Kawai, N. Tamai, M. Irie, J. Am. Chem. Soc. 2004, 126,
14843–14849.
[4] a) G. Szalꢄki, J. L. Pozzo, Chem. Eur. J. 2013, 19, 11124–11132; b) K. Bey-
doun, J. Roger, J. Boixel, H. Le Bozec, V. Guerchais, H. Doucet, Chem.
Commun. 2012, 48, 11951–11953.
[19] J. Massaad, J.-C. Micheau, C. Coudret, R. Sanchez, G. Guirado, S. Del-
baere, Chem. Eur. J. 2012, 18, 6568–6575. See also a) S. Delbaere, J. Ber-
thet, T. Shiozawa, Y. Yokoyama, J. Org. Chem. 2012, 77, 1853–1859;
Chem. Eur. J. 2015, 21, 1 – 13
www.chemeurj.org
11
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
b) ref. [6]; c) S. Alo¨ıse, M. Sliwa, G. Buntinx, S. Delbaere, A. Perrier, F.
Maurel, D. Jacquemin, M. Takeshita, Phys. Chem. Chem. Phys. 2013, 15,
6226–6234; d) O. Galangau, Y. Kimura, R. Kanazawa, T. Nakashima, T.
Kawai, Eur. J. Org. Chem. 2014, 7165–7173.
Vanthuyne, J. A. G. Williams, C. Lescop, C. Roussel, J. Autschbach, J.
Crassous, R. Rꢆau, Angew. Chem. Int. Ed. 2010, 49, 99–102; Angew.
Chem. 2010, 122, 103–106; d) J.-D. Chen, H.-Y. Lu, C.-F. Chen, Chem. Eur.
J. 2010, 16, 11843–11846; e) T. B. Norsten, A. Peters, R. McDonald, M.
Wang, N. R. Branda, J. Am. Chem. Soc. 2001, 123, 7447–7448; f) T. J. Wig-
glesworth, D. Sud, T. B. Norsten, V. S. Lekhi, N. R. Branda, J. Am. Chem.
Soc. 2005, 127, 7272–7273.
[20] a) Naraso, F. Wudl, Macromolecules 2008, 41, 3169–3174; b) M. Kariko-
mi, C. Kitamura, S. Tanaka, Y. Yamashita, J. Am. Chem. Soc. 1995, 117,
6791–6792.
[21] E. R. Johnson, S. Keinan, P. Mori-Sanchez, J. Contreras-Garcꢈa, A. J.
Cohen, W. Yang, J. Am. Chem. Soc. 2010, 132, 6498–6506.
[23] N. Ito, T. Hirose, K. Matsuda, Org. Lett. 2014, 16, 2502–2505.
[22] For general reviews on helicenes, see: a) Y. Shen, C.-F. Chen, Chem. Rev.
2012, 112, 1463–1535; b) T. Okuyama, Y. Tani, K. Miyake, Y. Yokoyama, J.
Org. Chem. 2007, 72, 1634–1638. See also: c) L. Norel, M. Rudolph, N.
Received: February 15, 2015
Published online on && &&, 0000
&
&
Chem. Eur. J. 2015, 21, 1 – 13
www.chemeurj.org
12
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
FULL PAPER
Simple substitutions: The preparation
of five new benzothiophene-centered
terarylene pro-helical derivatives is de-
scribed. They display various and oppo-
site photoactivities, depending on sub-
stitution at the reactive centers (see
figure). When fully substituted, ring cyc-
lization generates stable [5]helical deriv-
atives with outstanding absorption
properties.
&
Helical Structures
O. Galangau,* T. Nakashima, F. Maurel,
T. Kawai*
&& – &&
Substituent Effects on the
Photochromic Properties of
Benzothiophene-Based Derivatives
Chem. Eur. J. 2015, 21, 1 – 13
www.chemeurj.org
13
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ